Five reasons why Gen AI may not be your exam BFF
When used ethically, Generative AI (Gen AI) tools such as ChatGPT and Microsoft Copilot are helpful tools for organising, idea generation, and accessibility — especially during busy periods like exams.
However, to ensure you don’t end up plagiarising, breaching academic integrity, or, as Hermione Granger famously once said, even worse, expelled*, you must use Gen AI in line with the university’s guidance.
It can be confusing to know what to do, so here are five key points to remember, so you don’t get caught out.
Gen AI might have made that up.
Most AI models are trained to provide a ‘likely’ answer that looks true even though it is completely made up. This frequent occurrence is commonly known as AI hallucinations. Because of this, the ‘facts’ you take from AI could be wrong. It is also common for AI tools to generate ‘phantom citations’ — in other words, sources that do not actually exist. Always check the information and references that AI gives you to make sure they are real and genuinely relevant to your work.
Tip: When looking for reliable, high-quality academic information for your work, Library Search is the way to go.
Gen AI doesn’t sound human.
Sometimes AI tools can help break through writer’s block, for example, by asking it for some inspiration to get you started. But asking it to generate paragraphs that you then make minor changes to is a big mistake when it comes to uni work. We’re now so used to reading AI writing that it’s become easier to spot its signs, even without checkers. No one knows this better than the academic experts who mark your work for a living.
Tip: The human skills that Gen AI cannot provide — original thinking, engagement with your course, factual accuracy, critical analysis — are what get you the grades.
Gen AI produces basic and generic answers.
Imagine this: you ask ChatGPT for a couple of paragraphs to help you answer a difficult exam question. But it turns out three other people in your seminar group did the same thing, and you’ve now unknowingly submitted the exact same essay points and references. Then the plagiarism accusations start…
That may be an exaggeration, but the point is, AI will never be able to do what you can: come up with a unique response to an academic challenge. Your ability to do that is what will set you apart in exams and when looking for jobs after university.
Tip: If you need help with academic writing, Skills@library has lots of advice.
Gen AI misuse breaks the university’s plagiarism rules.
This is a big one. There are clear rules regarding AI use and plagiarism. Breaking them, even accidentally, is a breach of academic integrity. If you present text, data, images, designs, sound, or performance from AI as your own, you will face a penalty for plagiarism.
Tip: The university has a traffic light system that students should use when deciding whether to use AI in assessments, so check it when in doubt.
Gen AI use can be unethical.
There are several ethical issues to be aware of when considering using AI; it’s down to you to weigh up the harms. Firstly, AI systems are energy-intensive and negatively impact our environment. Secondly, AI systems are questionable when it comes to privacy, as they scrape information from across the internet, regardless of whether you have consented to this or not. Thirdly, studies show that biases and racism are built into AI algorithms, reflecting the discrimination many people experience in society.
So next time you go to type that Chat GPT prompt, ask yourself if you could get that answer elsewhere.
Read more about how to use AI properly at university on the Gen AI hub.
